Research ArticleThe Efficacy and Safety of TranscatheterArterial Embolization to Treat Renal Hemorrhage afterPercutaneous Nephrolithotomy

Nan Du1,2, Jing-QinMa1, Jian-Jun Luo2, Qing-Xin Liu1, Zi-Han Zhang1, Min-Jie Yang1,Tian-Zhu Yu1, Yun Tao1, Rong Liu1,2, Wen Zhang 1,2, and Zhi-Ping Yan 1,2

1Department of Interventional Radiology, Zhongshan Hospital, Fudan University, Shanghai 200032, China2Shanghai Institute of Medical Imaging, Shanghai 200041, China

Correspondence should be addressed to Wen Zhang; zhang.wen2@zs-hospital.sh.cn andZhi-Ping Yan; yan.zhiping@zs-hospital.sh.cn

Received 16 January 2019; Revised 30 March 2019; Accepted 3 April 2019; Published 22 April 2019

Academic Editor: Yoshito Tsushima

Copyright © 2019 Nan Du et al.This is an open access article distributed under the Creative Commons Attribution License, whichpermits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Purpose. The aim of this study was to evaluate the safety and efficacy of transcatheter arterial embolization (TAE) in patients withrenal hemorrhage after percutaneous nephrolithotomy (PCNL) and evaluate the risk factors that may result in severe bleedingrequiring TAE.Methods. We retrospectively reviewed 121 patients with post-PCNL renal hemorrhage.Thirty-two patients receivingendovascular embolization were compared with 89 patients only receiving conservative treatment. The demographic and clinicaldata were recorded and compared between the two groups. The values of estimated glomerular filtration rate (eGFR) and serumcreatinine (SCr) were recorded preoperatively, postoperatively, and at last follow-up and analyzed to evaluate the safety andefficiency of TAE. Results. The successful hemostasis rate of conservative therapy was 73.6% (89/121) and that of TAE was 100%(32/32). SCr and eGFR were not significantly different before PCNL and after the last follow-up of TAE (SCr: 0.95 vs. 0.95 mg/dl,P=0.857; eGFR: 86.77 vs. 86.18 ml/min/1.73m2, P=0.715). The univariate analysis demonstrated that advanced age, urinary tractinfection, and diabetes mellitus were significantly associated with severe bleeding during PCNL. Multivariate analysis furtheridentified that diabetes mellitus was an independent risk factor for severe bleeding needing TAE [odds ratio (OR): 3.778, 95%confidence interval (CI):1.276-11.190, and P=0.016]. Conclusion. TAE is a safe and effective procedure to treat renal hemorrhagethat cannot be resisted by conservative treatment after PCNL. Diabetes mellitus was associated with high risks of severe bleedingneeding TAE after PCNL.

1. Introduction

Percutaneous nephrolithotomy (PCNL) is a safe and effectivetreatment for patients with upper urinary calculi. Hematuriais a major and life-threatening complication of PCNL whichrequires promptmanagement [1–3]. Previous study indicatedthat transfusion rates vary from 1% to 11% and embolizationrates vary from 0.6% to 2.6% after PCNL [4–6]. Beforeembolization, angiographywas often performed for diagnosisof iatrogenic renal arterial lesions [2, 7, 8]. Moreover, tran-scatheter arterial embolization (TAE) has been regarded asan effective method for postoperative hemorrhage of PCNL[7, 9]. Before receiving interventional therapy, patients often

undergo conservative treatment to hemostasis, includingnephrostomy tube clamping, adequate hydration, hemostaticdrug using, and blood transfusion [10–12]. Patients withhemodynamic instability or refractory bleeding who cannotachieve hemostasis by conservative therapy have been rec-ommended to angiographic embolization for diagnosis andtreatment [11, 13]. Although the efficacy of TAE for post-PCNL hemorrhage has been previously explored [11, 14], thelong-term effects on renal function remain to be furtherconfirmed. However, existing studies have not directly com-pared patients undergoing TAE with patients only receivingconservative treatment. The purpose of this study was toevaluate the safety and efficacy of TAE for patients with

HindawiBioMed Research InternationalVolume 2019, Article ID 6265183, 10 pageshttps://doi.org/10.1155/2019/6265183

2 BioMed Research International

post-PCNL severe hemorrhage. Furthermore, we comparedpatients receiving TAEwith patients treated conservatively toexplore the risk factors associated with severe hemorrhage.

2. Methods

2.1. Patients Selection. From October 2010 to June 2017, weretrospectively reviewed 812 patients who underwent PCNLprocedures. A total of 121 patients (14.9%), including 89cases who received conservative treatment and 32 cases whoreceived TAE, were enrolled in this study. The indications forangiography and embolization in our hospital included signsindicative of vascular injury (such as hemodynamic insta-bility requiring blood transfusion, nonresolving hematuriain hemodynamically stable patients, and nonremitting flankpain); uncontrolled intraoperative blood loss; and imagingevidence of vascular injury (pseudoaneurysm formation,active extravasation of contrast medium, creation of arteri-ovenous fistula, etc.). This study was approved by the ethicscommittee and institutional review board of our institution.All patients signed informed consent before interventionalangiography and embolization.

2.2. Data Collection. Demographic and clinical detailsrecorded included age, sex, body mass index (BMI), stonesize, side and location, medical comorbidities, and historyof antiplatelet therapy and anticoagulation therapy. Types oflesions on angiography were classified into contrast mediumextravasation, pseudoaneurysm (PA), arteriovenous fistula(AVF), arteriocalyceal fistula (ACF), or a complex lesion.The levels of hemoglobin (HB) were recorded at preoperativePCNL, postoperative PCNL, and before the day of TAE.Blood loss was defined as HB drop between preoperativePCNL and the day before conservative treatment/TAE.Other laboratory data accessed included platelet count, C-reactive protein (CRP), procalcitonin, urine white blood cellcount, and the levels of eGFR and SCr before TAE, afterembolization, and at last follow-up. Urinary tract infectionwas diagnosed by abnormal laboratory data of the leukocytecount in blood and urine, the level of neutrophil, and CRP.

The TAE success and complications were independentlyevaluated. TAE success was defined as the absence of clinicaland imaging evidence of further bleeding within 1 weekand stable renal function. Imaging examination, such asultrasonography and computed tomography (CT) imaging,was performed during the follow-up time. Days from post-PCNLhemorrhage to discharge were recorded as hospitaliza-tion days and compared between the two groups. Whetherpatients received transfusion during hospitalization and theunits of transfusion were recorded.

2.3. PCNL Procedures. All patients underwent PCNL in theguidance of fluoroscopy. Percutaneous access to the calyx wasachieved under general anesthesia. After the tract was dilatedwith nephrostomydilator set, an 18-FrAmplatz sheath (Cook,Inc., Bloomington, USA) was inserted into the targeted calyx.Nephroscopy was performed with a Wolf Nephroscope, andthe stones were fragmented using a pneumatic lithotripteror holmium laser. The stone fragments were removed with

irrigation or grasping forceps. At the end of the procedure,a ureteral stent and a nephrostomy tube were left to ensurepostoperatively adequate urine drainage. If there was nocomplication or significant residual stone, the nephrostomytube was removed after 3 to 5 days, and the nephrostomytube was removed after 2 to 4 weeks. Patients were closelymonitored to exclude bleeding. According to the extent ofbleeding, patients received conservative therapy or furtherTAE.

2.4. TAE Procedures. Transfemoral catheterization was per-formed under local anesthesia. After vascular access wasobtained, a 4- or 5-French pigtail catheter (Cook, Inc.,Bloomington, IN) was placed in the abdominal aorta. Dig-ital subtraction angiography (DSA) using Visipaque (GEHealthcare, Cork, Ireland) was performed to evaluate therenal arteries, accessory renal arteries, lumbar arteries, andother collateral vessels. Then, angiography with a 4- or 5-Fr Cobra catheter (Cook, Inc.) was performed to evaluatethe renal artery conditions and the location of the lesions.A 2.8-Fr Renegade catheter (Boston Scientific, Natick, MA)was used for superselective embolization of the culprit arteryas distally as possible to minimize the loss of normal renalparenchyma. Embolization materials such as coils (Cook,Inc.), gelatin sponge (Jingling, Jiangsu, China), and n-butyl-2-cyanoacrylate (NBCA) (Compont, Beijing, China) wereused according to the types of the lesions. After embolization,angiogram was performed to verify the occlusion of theculprit arteries and to assess the percent of parenchymal loss(infarction area). All the patients were carefully monitoredafter the procedure to exclude potential rebleeding.

3. Statistical Analysis

Statistical analysis was performed using SPSS 22.0 (SPSS,Chicago, Illinois). Categorical variables were presented asnumbers with percentage, and continuous variables werepresented as mean ± standard deviation (SD).The differencesbetween the groups were evaluated using the Chi-square testor Fisher’s exact test for categorical variables and the Studentt-test or Mann-Whitney U test for continuous variables.Multivariate binary logistic regressionwas used formultivari-ate analysis. Odds ratio (OR) with 95% confidence interval(CI) was used to measure the magnitude of association. Atwo-sided P value <0.05 was considered to be statisticallysignificant.

4. Results

In the present study, 3.9% (32 cases) received TAE for severebleeding after PCNL. Before TAE, the conservative therapyincluded drainage tube clamping, hemostatic drugs, antibi-otics, adequate hydration, blood transfusion, and absolutebed rest. If bleeding persisted (continuing or intermittenthematuria that led to a further drop in hematocrit, tachy-cardia, and hypotension), selective renal arteriography withembolization was used to stop the bleeding. The successfulhemostasis rate of conservative therapy was 73.6% (89/121)and that of TAE was 100% (32/32). The demographic and

BioMed Research International 3

(a) (b)

Figure 1: (a) Selective renal arteriography shows a PA arising from the right inferior anterior interlobar artery in a 49-year-old woman withrenal hemorrhage 5 days after PCNL. (b) After finding the responsible vessel and embolized with NBCA and coils, angiographic imagingshows the disappearance of the PA.

clinical data between the two groups were presented inTable 1. The mean platelet counts before PCNL had nodifference between the two groups (P=0.228). No differencewas found between the two groups about patients receivinganticoagulation or antiplatelet therapy. Univariate analysisindicated that advanced age (P=0.039), preoperative urinarytract infection (P=0.024), and diabetes mellitus needingembolization (P=0.001) were associated with severe postop-erative bleeding. Stone size and types did not increase therisk of severe bleeding needing TAE (P=0.483 and P=0.170).The location of calyx entry had no difference between thetwo groups (P=0.246). There was a significantly higherblood loss for patients who received TAE than for patientswho only received conservative therapy (58.3 vs. 15.3 g/l,P<0.001). In the TAE group, 56.3% of patients receivedtransfusion, and the mean volume was 6.7 units, whichwere significantly higher than the conservative treatmentgroup (56.3% vs. 22.5%, P<0.001; 6.7 vs. 2.7, P=0.002). Andpatients had prolonged hospitalization days in the TAE groupcompared with the control group (7.5 days vs. 4.1 days,P<0.001).

The potential risk factors, including age, stone size, stonetype, diabetes mellitus, urinary tract infection, and puncturesite, were entered into multivariate analysis to identify theindependent factors for post-PCNL severe bleeding. Resultsof multivariate analysis were shown in Table 2. Multivariatelogistic regression analysis identified that diabetes melli-tus was independently associated with severe postbleedingrequiring TAE (OR: 3.778; 95%CI: 1.276-11.190; and P=0.016).

The operation information during TAE was summarizedin Table 3. The angiographic imaging showed that 3 patientswith vascular injuries located in upper renal pole, 7 patientswith vascular injuries located inmiddle renal pole, 20 patientswith vascular injuries located in in lower renal pole, and2 patients with vascular injuries located in middle andlower renal pole. The following was shown: 14 patients

with contrast medium extravasation, 10 patients with PA, 4patients with AVF, and 4 patients with composite lesions.All of the patients had successfully stopped bleeding andsome representative cases were depicted in Figures 1–4. Theembolic material used in TAE was coil in 27 patients, coil andgelatin sponge in 4 patients, and coil combined with NBCAin 1 patient. The mean number of coils used was 3 (range,1 to 12 coils). The angiogram was performed to evaluatethe renal parenchymal loss, and there was no patient withparenchymal loss more than 25% of the total renal volume.One patient was successfully controlled by receiving a secondTAE for rehemorrhage. Also, in the 8 patients with solitaryor functional solitary kidney, 1 received TAE and 7 onlyreceived conservative treatment. They were all followed upintensively, and no further deterioration of renal function wasobserved. Endovascular embolization as distant as possiblewas performed in all 32 patients to reduce the loss of kidneyparenchymal.

The mean time of follow-up was 18.6 months (range,12 to 36 months) in all patients. No major complicationsof embolization (renal artery dissection, coil migration,postembolization syndrome, and loss of renal function)occurred in patients received TAE. The values of SCr andeGFR in different time points were shown in Table 4 (Patientswith eGFR more than 120 ml/min/1.73m2 were recordedas 120 ml/min/1.73m2 .) No statistical difference of SCr wasobserved between the two groups at preoperative PCNLand at last follow-up (0.95 vs. 0.91 mg/dl, P=0.446 and0.95 vs. 0.92 mg/dl, P=0.635) and there was no significantdifference in eGFR (86.77 vs. 88.87 ml/min/1.73m2 , P=0.552and 86.18 vs. 85.11, P=0.770). Moreover, no differences of SCrand eGFR were found before PCNL and at last follow-upin the TAE group (0.95 vs. 0.95 mg/dl, P=0.857 and 86.77vs. 86.18 ml/min/1.73m2, P=0.715), and also no differenceswere found in the conservative treatment group (0.91 vs.0.92 mg/dl, P=0.698 and 88.87 vs. 85.11 ml/min/1.73m2 ,P=0.125).

4 BioMed Research International

Table 1: Demographic and clinical difference between patients in the TAE group and conservative treatment group.

Characteristic All TAE group Control group P value(n=121) (n=32) (n=89)

Age 51.9 ± 11.8 55.6 ± 9.6 50.6 ± 12.3 0.039∗Gender (male) 0.649

Male 87(71.9) 24(75.0) 63(70.8)Female 34(28.1) 8(25.0) 26(29.2)

Platelet count (×10∧9/L) 203.7 ± 53.4 193.7 ± 55.0 207.3 ± 52.6 0.228Stone size 0.483<2 44(36.4) 10(31.3) 34(38.2)≥2 cm 77(63.6) 22(68.7) 55(61.8)

Stone type 0.170Solitary 34(28.1) 6(18.8) 28(31.5)Multiple or Staghorn 87(71.9) 26(81.2) 61(68.5)

Number of tracts 0.394Simple 116(95.9) 32(100.0) 84(94.4)Multiple 5(4.1) 0(0) 5(5.6)

Urinary tract infection 0.024∗Yes 55(45.5) 20(62.5) 35(39.3)No 66(54.5) 12(37.5) 54(60.7)

Hypertension 0.436Yes 46(38.0) 14(43.8) 32(36.0)No 75(62.0) 18(56.2) 57(64.0)

Diabetes mellitus 0.001∗Yes 22(18.2) 12(37.5) 10(11.2)No 99(81.8) 20(62.5) 79(88.8)

Antiplatelet therapy 0.567Yes 7(5.8) 3(9.4) 4(4.5)No 114(94.2) 29(90.6) 85(95.5)Anticoagulation therapy 1.000Yes 4(3.3) 1(3.1) 3(3.4)No 117(96.7) 31(96.9) 86(96.6)History of operation 0.831

None 67(55.4) 18(56.3) 49(55.1)ESWL 13(10.7) 4(12.5) 9(10.1)PCNL and/or POS 41(33.9) 10(31.2) 31(34.8)

Operation site 0.260Right 54(44.6) 17(53.1) 37(41.6)Left 67(55.4) 15(46.9) 52(58.4)

Puncture site 0.246Upper calyx 17(14.0) 5(15.6) 12(13.5)Middle calyx 64(52.9) 13(40.6) 51(57.3)Inferior calyx 40(33.1) 14(43.8) 26(29.2)

Hemoglobin decrease (g/L) 26.4 ± 24.6 58.3 ± 18.5 15.3 ± 14.7 <0.001∗Numbers of patients receiving transfusion <0.001∗

Yes 38(31.4) 18(56.3) 20(22.5)No 83(68.6) 14(43.8) 69(77.5)

Units of transfusion 6.7 ± 4.4 2.7 ± 1.5 0.001∗Duration of hospitalization (days) 7.5 ± 4.1 4.2 ± 1.5 <0.001∗ESWL: extracorporeal shock wave lithotripsy; PCNL: percutaneous nephrolithotomy; POS: previous ipsilateral open renal surgery. ∗P<0.05.

BioMed Research International 5

(a) (b)

(c) (d)

Figure 2: (a)The angiographic imaging in a 56-year-oldman with renal hemorrhage 3 days after PCNL indicates a PA arise from the superioranterior interlobar artery (thin arrow). (b) Angiographic imaging show that an AVF (thick arrowhead) arise from the posterior and superioranterior interlobar artery. (c, d) Angiogram after coil embolization demonstrates complete occlusion of the PA and AVF.

Table 2: Multivariate logistic analysis of the potential factors for severe bleeding needing TAE.

Factors Characteristics OR 95% CI P valuesUnfavorable Favorable

Age 1.027 0.984-1.073 0.224Stone size ≥2cm <2cm 1.276 0.471-3.452 0.632Stone type Multiple or Staghorn Solitary 2.614 0.825-8.276 0.102Urinary tract infection Yes No 2.063 0.806-5.283 0.131Diabetes mellitus Yes No 3.778 1.276-11.190 0.016∗Puncture site Inferior calyx Middle or upper calyx 1.905 0.742-4.893 0.180OR: odds ratio; CI: confidence interval. ∗P<0.05.

5. Discussion

PCNL is an efficient way for kidney stones and ureteral stones[15, 16]. Bleeding is one of the most common clinical compli-cations of PCNL. In most patients, post-PCNL hemorrhagecould be controlled by conservative therapy. However, TAEhas been an established therapy for severe, persistent, orintermittent hemorrhage after PCNL that cannot be stoppedby conservative treatment [17, 18]. In previous literatures,large stone burdens with multiple punctures, lower renal

puncture site, stone number and types, history of ipsilateralrenal stone surgery, and occurrence of intraoperative pelvica-lyceal perforation have been demonstrated as the risk factorsfor severe post-PCNL hemorrhage [19–22]. Moreover, estab-lished studies showed that patients with tubeless drainageafter PCNL were also associated with an increased risk ofbleeding and subsequent angioembolization [22]. However,therewere various datawith regards to the risk factors of post-PCNL severe bleeding, which needs to be further confirmed.In this study, we compared patients who received TAE with

6 BioMed Research International

(a) (b)

(c) (d)

Figure 3: (a) Selective renal arteriogram shows a PA arise from the left posterior anterior segmental artery in a 54-year-old man. (b) Renalarteriography during the arterial phase shows a PA (thin arrow) and contrast agent extravasation along the drainage pipe (thick arrowhead)arising from the posterior and superior anterior interlobar artery. (c) After embolization with 3 coils (TORNADO MWCE-18S-3.0-2, CookInc.), there is no contrast medium extravasation again under angiogram. (d) After embolization, digital subtraction angiographic imagingshows complete occlusion of the PA and contrast agent retention.

patients who only received conservative therapy to evaluatethe efficiency of endovascular embolization and the potentialrelated risk factors.

In our study, the blood transfusion rate was 4.7% (38/812)for patients with post-PCNL hemorrhage. The embolizationrate after PCNL in our hospital was 3.9% (32/812). In univari-ate analysis, advanced age, preoperative urinary tract infec-tion, and diabetes mellitus needing embolization were shownto be associatedwith post-PCNL severe bleeding. For patientswith advanced age, the ability of repairing injury declined,which could result increased blood losing. For preoperativeurinary tract infection, the possible explanations includedthe hyperemic nature of inflamed urothelium, or distortedanatomy secondary to edema. However, advanced age andpreoperative urinary tract infection did not reveal statisticalsignificance inmultivariate analysis.Thismay be explained bythe small number of patients included in our study. Diabetesmellitus was an independent predictive factor affecting thebleeding risk, which agree with the study of Akman et al.and Kurtulus et al. [23, 24].The relationship between diabetesand bleeding after the initial trauma of tract formation wasexplained with arteriosclerosis. In addition, diabetes affects

the whole vascular system, resulting in microangiopathies,which are highly vulnerable to bleeding [25].

The stone size and types did not increase the risk ofpost-PCNL severe bleeding (P=0.483 and P=0.170). Also,the different puncture site location had no correlation withincreased bleeding risk (P=0.246). Patients with history ofurinary system operation were recorded as extracorporealshock-wave lithotripsy (ESWL), PCNL, and previous ipsilat-eral open renal surgery (POS), and no evidence indicatedincreased risk of bleeding. The blood loss and transfu-sion units of patients who received TAE were obviouslyhigher than patients who only received conservative therapy.Therefore, patients receiving TAE might suffer severe renalvascular injury and severe bleeding, and they might be morerecommended to angiographic embolization.

Arterial lesions are rare but often ineffectively treatedby conservative means. Owing to the high pressure, renalarterial bleeding can drain into renal parenchyma or hilarareolar tissue resulting in a PA [3, 26]. In accordance withprevious studies, contrast medium extravasation and PAwere the most common types of angiographic manifesta-tions [10, 11, 27]. In previous study, the major limitations

BioMed Research International 7

Figure 4: Secondary embolization in a patient with PA after PCNL. (a) Abdominal aorta angiography cannot show the number of left renalartery clearly because of the interference of intestinal gas. (b, c) Selective left renal arteriography shows a PA (red arrow). (d) Angiogram aftercoil embolization showing complete occlusion of the PA. (e-h) Angiogram 3-day after first time TAE shows a new PA (red arrow) arising fromaccessory left renal artery, which arises at approximately the inferior level of third Lumbar vertebra. (i) After coil embolization, angiogramshows complete occlusion of the PA. (j, k) The CT image shows the location of left renal artery and accessory left renal artery.

Table 3: Operation information about TAE groups.

Characteristic Number Percent (%)Site of vascular lesions (n)

Upper pole 3 9.4Middle pole 7 21.9Lower pole 20 62.5Middle & lower pole 2 6.2

Angiographic findings (n)Contrast extravasation 14 43.7PA 10 31.3AVF 4 12.5PA & AVF 2 6.3PA & ACF 1 3.1PA & AVF & AVF 1 3.1

Embolic material used (n)Coil 27 84.4Coil & GS 4 12.5Coil & Glue 1 3.1

Parenchymal loss (%)Less than 25% 32 100

PA: pseudoaneurysm; AVF: renal arteriovenous fistula; ACF: arteriocalycealfistula; and GS: gelatin sponge.

of arterial embolization with NBCA, gelatin sponge, PVA,and Embosphere include the reflux of the embolic agentto the normal artery, which leads to unwanted regionalembolization; passing to the venous system from AVF, whichleads to pulmonary embolism; passing to the collectingsystem from ACF, which leads to acute urinary obstruction;and inadequacy in occluding large vessels injury [28]. In ourexperience, coils, as a permanent embolic agent, were aneffective renal arterial occlusion tool, particularly in patientswith small artery lesions.

No complication of coil migration was found in long-term follow-up. In previous literature, the efficacy of TAEin treating post-PCNL hemorrhage was 88%-100% [7, 29].However, ischemia in renal parenchyma is one problemof TAE that may affect the renal function [29]. Moreover,the long-time effects of contrast medium on renal functionshould be considered as well. In this study, the ultrasonogra-phy or CT imaging in postoperative follow-up showed thatthe shrinking of the infarct area and the development ofcollaterals were found in TAE patients. No recanalizationor migration of coil was found in long-term follow-up. Nopatients with loss of parenchymal larger than 25% of the totalrenal volume were found, and renal functions were effectivelyreserved in long-term follow-up.Thismay owe to the effective

8 BioMed Research International

Table 4: The comparison of laboratory examination data perioperative and long-term renal function outcomes between groups.

Variable TAE group Control group P values(n=32) (n=89)

Preoperative renal functionSCr (mg/dl) 0.95 ± 0.26 0.91 ± 0.23 0.446eGFR (ml/min/1.73m2) 86.77 ± 18.76 88.87 ± 15.99 0.552

Postoperative renal functionSCr 0.94 ± 0.25 0.93 ± 0.22 0.761eGFR (ml/min/1.73m2) 87.60 ± 20.24 86.41 ± 18.95 0.775

Renal function at lost follow-upSCr (mg/dl) 0.95 ± 0.30 0.92 ± 0.28 0.635eGFR (ml/min/1.73m2) 86.18 ± 15.78 85.11 ± 16.95 0.770

Renal function change between preoperational and at lost follow-upSCr (P value) 0.857 0.698eGFR (P value) 0.715 0.125

eGFR: estimated glomerular filtration rate; SCr: serum creatinine; and TAE: transcatheter arterial embolization. ∗P<0.05.

compensation of contralateral normal kidney. In all, TAE wasan effective method to resolve post-PCNL severe hemorrhagewithout obvious side effect on renal function.

From this study and our experience, we proposed 3aspects that should be given special attention. First, TAE isa safe and effective procedure to control post-PCNL severebleeding without deterioration of renal function, even forpatients receiving repeated TAE. Second, patients, who com-plicated with persisted or intermittent hematuria requiringmultiple blood transfusion, can receive early angiogram andTAE to reduce the hospitalization days and the unit ofblood transfusion. Third, to avoid the risks of post-PCNLsevere bleeding, surgeons should operatemore prudently andcarefully on patients with diabetes mellitus.

The limitations of the present study were its retrospectivenature, which introduced selection biases. Even though TAEhas been regarded as a standard method for severe bleedingafter PCNL, there is no criteria consensus about how andwhen to intervene by interventional therapy. Hence, theultimate decision to use TAE was made on a case-by-casebasis by the operator, which introduced additional selectionbias. Lack of standardized criteria to evaluate the loss of renalparenchymal introduces another potential bias.

6. Conclusions

Severe renal hemorrhage is a rare but serious complicationof PCNL. TAE is a safe and effective minimally invasivetreatment for renal hemorrhage after PCNL that cannot becontrolled by conservative therapy. Early TAE may decreasethe units of blood transfusion and hospitalization daysfor patients. Selective catheterization and embolization canhelp reserve renal function and reduce the potential riskof nephrectomy. Moreover, patients with diabetes mellitusare correlated with an increased risk of post-PCNL renalhemorrhage needing TAE.

Abbreviations

PCNL: Percutaneous nephrolithotomyTAE: Transcatheter arterial embolizationeGFR: Estimated glomerular filtration ratePA: PseudoaneurysmAVF: Arteriovenous fistulaACF: Arteriocalyceal fistulaCT: Computed tomographyHB: HemoglobinCRP: C-reactive proteinSCr: Serum creatinineDSA: Digital subtraction angiographyNBCA: n-Butyl-2-cyanoacrylateESWL: Extracorporeal shock-wave lithotripsyOR: Odds ratioCI: Confidence interval.

Data Availability

The data used to support the findings of this study areavailable from the corresponding author upon appropriaterequest.

Ethical Approval

This study was approved by the ethics committee of Zhong-shan Hospital, Fudan University, Shanghai, China.

Conflicts of Interest

The authors declare no conflicts of interest.

Authors’ Contributions

Nan Du and Jing-Qin Ma should be the first authors forcollecting and analyzing data and writing this manuscript.They made the same contributions to the manuscript. Wen

BioMed Research International 9

Zhang and Zhi-Ping Yan, as the corresponding authors, madethe major contributions to design and conduct of this study.Zi-Han Zhang, Min-Jie Yang, Tian-Zhu Yu, and Yun Taomade efforts to collect data and statistical analyses. Jian-JunLuo, Qing-Xin Liu, and Rong Liu provided the study patients.All authors read and approved the final manuscript. Nan Duand Jing-Qin Ma contributed equally to this work.

Acknowledgments

This work was supported by the Project of Scientific Researchform Science and Technology Commission of ShanghaiMunicipality (16411968600, 17YF1417500). We should giveour thanks to the Departments of Medical Record and Inter-ventional Radiology for offering data records.The authors arevery grateful to Dr. Xianglin Hu in Zhongshan Hospital ofFudan University for his very professional suggestions aboutEnglish writing.

References

[1] C. M. Sommer, U. Stampfl, N. Bellemann et al., “Patients withlife-threatening arterial renal hemorrhage:CT angiography andcatheter angiography with subsequent superselective emboliza-tion,” CardioVascular and Interventional Radiology, vol. 33, no.3, pp. 498–508, 2010.

[2] M. Canstasdemir, I. Adaletli, D. Cebi, F. Kantarci, N. D.Selcuk, and F. Numan, “Emergency endovascular embolizationof traumatic intrarenal arterial pseudoaneurysms with N-Butylcyanoacrylate,” Clinical Radiology, vol. 58, no. 7, pp. 560–565,2003.

[3] L. Richstone, E. Reggio, M. C. Ost et al., “First Prize (tie): hem-orrhage following percutaneous renal surgery: characterizationof angiographic findings,” Journal of Endourology, vol. 22, no. 6,pp. 1129–1135, 2008.

[4] M. M. El Tayeb, J. J. Knoedler, A. E. Krambeck, J. E. Paonessa,M. J. Mellon, and J. E. Lingeman, “Vascular complicationsafter percutaneous nephrolithotomy: 10 years of experience,”Urology, vol. 85, no. 4, pp. 777–781, 2015.

[5] C. Seitz, M. Desai, A. Hacker et al., “Incidence, prevention,and management of complications following percutaneousnephrolitholapaxy,”EuropeanUrology, vol. 61, no. 1, pp. 146–158,2012.

[6] S. R. Keoghane, R. J. Cetti, A. E. Rogers, and B. H. Walmsley,“Blood transfusion, embolisation and nephrectomy after percu-taneous nephrolithotomy (PCNL),” BJU International, vol. 111,no. 4, pp. 628–632, 2013.

[7] K. Sam, G. Gahide, G. Soulez et al., “Percutaneous embolizationof iatrogenic arterial kidney injuries: Safety, efficacy, and impacton blood pressure and renal function,” Journal of Vascular andInterventional Radiology, vol. 22, no. 11, pp. 1563–1568, 2011.

[8] A. N. Keeling, F. P. McGrath, J. Thornton, P. Brennan, and M.J. Lee, “Emergency percutaneous transcatheter embolisation ofacute arterial haemorrhage,” Irish Journal of Medical Science,vol. 179, no. 3, pp. 385–391, 2010.

[9] E. Mavili, H. Donmez, N. Ozcan, M. Sipahioglu, and A. Demir-tas, “Transarterial embolization for renal arterial bleeding,”Diagnostic and Interventional Radiology, vol. 15, pp. 143–147,2009.

[10] V. Jinga, B. Dorobat, S. Youssef et al., “Transarterial emboliza-tion of renal vascular lesions after percutaneous nephrolitho-tomy,” Chirurgia (Romania), vol. 108, no. 4, pp. 521–529, 2013.

[11] L. Li, Y. Zhang, Y. Chen et al., “Amulticentre retrospective studyof transcatheter angiographic embolization in the treatmentof delayed haemorrhage after percutaneous nephrolithotomy,”European Radiology, vol. 25, no. 4, pp. 1140–1147, 2015.

[12] Y. Zhao, B. Zhong, S. Xu, B. Shen, and F. Tan, “Interven-tional treatment of massive hemorrhage after percutaneousnephrolithotomy,” National Medical Journal of China, vol. 94,no. 30, pp. 2369–2371, 2014.

[13] H. L. Wang, C. Y. Xu, H. H. Wang, and W. Xu, “Emergencytranscatheter arterial embolization for acute renal hemorrhage,”Medicine (United States), vol. 94, no. 42, p. e1667, 2015.

[14] T. Ząbkowski, P. Piasecki, H. Zielinski, A. Wieczorek, K. Brzo-zowski, and P. Zięcina, “Superselective renal artery emboliza-tion in the treatment of iatrogenic bleeding into the urinarytract,”Medical Science Monitor, vol. 21, pp. 333–337, 2015.

[15] W. J. Wu and Z. Okeke, “Current clinical scoring systemsof percutaneous nephrolithotomy outcomes,” Nature ReviewsUrology, vol. 14, no. 8, pp. 459–469, 2017.

[16] M. S. Borofsky and J. E. Lingeman, “The role of open andlaparoscopic stone surgery in the modern era of endourology,”Nature Reviews Urology, vol. 12, no. 7, pp. 392–400, 2015.

[17] N. Poyraz, M. Balasar, I. E. Gokmen et al., “Clinical efficacy andsafety of transcatheter embolization for vascular complicationsafter percutaneous nephrolithotomy,” Videosurgery and OtherMiniinvasive Techniques/ Wideochirurgia i Inne Techniki MaloInwazyjne, vol. 12, no. 4, pp. 403–408, 2017.

[18] B. K. Somani, G. Nabi, P. Thorpe, and S. McClinton, “Endovas-cular control of haemorrhagic urological emergencies: anobservational study,” BMC Urology, vol. 6, no. 27, 2006.

[19] Y. Wang, Y. Wang, F. Jiang et al., “Post-percutaneous nephro-lithotomy septic shock and severe hemorrhage: a study of riskfactors,” Urologia Internationalis, vol. 88, no. 3, pp. 307–310,2012.

[20] G. Zeng, Z. Zhao, S. Wan et al., “Failure of initial renal arterialembolization for severe post-percutaneous nephrolithotomyhemorrhage: a multicenter study of risk factors,”The Journal ofUrology, vol. 190, no. 6, pp. 2133–2138, 2013.

[21] S. H. A. Said, M. A. Al Kadum Hassan, R. H. G. Ali, I.Aghaways, F. H. Kakamad, and K. Q. Mohammad, “Percuta-neous nephrolithotomy; alarming variables for postoperativebleeding,” Arab Journal of Urology, vol. 15, no. 1, pp. 24–29, 2017.

[22] A. Nouralizadeh, S. A. Ziaee, S. H. Hosseini Sharifi etal., “Delayed postpercutaneous nephrolithotomy hemorrhage:prevalence, predictive factors and management,” ScandinavianJournal of Urology, vol. 48, no. 1, pp. 110–115, 2014.

[23] F. Kurtulus, A. Fazlioglu, Z. Tandogdu, S. Karaca, Y. Salman,andM. Cek, “Analysis of factors related with bleeding in percu-taneous nephrolithotomy using balloon dilatation,” CanadianJournal of Urology, vol. 17, no. 6, pp. 5483–5489, 2010.

[24] T. Akman, M. Binbay, E. Sari et al., “Factors affecting bleedingduring percutaneous nephrolithotomy: single surgeon experi-ence,” Journal of Endourology, vol. 25, no. 2, pp. 327–333, 2011.

[25] A. Tefekli, H. Kurtoglu, K. Tepeler et al., “Does the metabolicsyndromeor its components affect the outcomeof percutaneousnephrolithotomy?” Journal of Endourology, vol. 22, no. 1, pp. 35–40, 2008.

[26] S. H. Mousavi-Bahar, S. Mehrabi, and M. K. Moslemi, “Per-cutaneous nephrolithotomy complications in 671 consecutive

10 BioMed Research International

patients: a single-center experience,”Urology Journal, vol. 8, no.4, pp. 271–276, 2011.

[27] J. D. Villalta, M. D. Sorensen, J. C. Durack, R. K. Kerlan, andM.L. Stoller, “Selective arterial embolization of angiomyolipomas:a comparison of smaller and larger embolic agents,”The Journalof Urology, vol. 186, no. 3, pp. 921–927, 2011.

[28] S. Guneyli, M. Gok, H. Bozkaya et al., “Endovascular manage-ment of iatrogenic renal arterial lesions and clinical outcomes,”Diagnostic and Interventional Radiology, vol. 21, no. 3, pp. 229–234, 2015.

[29] A. R. EL-Nahas, A. A. Shokeir, T.Mohsen et al., “Functional andmorphological effects of postpercutaneous nephrolithotomysuperselective renal angiographic embolization,” Urology, vol.71, no. 3, pp. 408–412, 2008.

Stem Cells International

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

MEDIATORSINFLAMMATION

of

EndocrinologyInternational Journal of

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Disease Markers

Hindawiwww.hindawi.com Volume 2018

BioMed Research International

OncologyJournal of

Hindawiwww.hindawi.com Volume 2013

Hindawiwww.hindawi.com Volume 2018

Oxidative Medicine and Cellular Longevity

Hindawiwww.hindawi.com Volume 2018

PPAR Research

Hindawi Publishing Corporation http://www.hindawi.com Volume 2013Hindawiwww.hindawi.com

The Scientific World Journal

Volume 2018

Immunology ResearchHindawiwww.hindawi.com Volume 2018

Journal of

ObesityJournal of

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Computational and Mathematical Methods in Medicine

Hindawiwww.hindawi.com Volume 2018

Behavioural Neurology

OphthalmologyJournal of

Hindawiwww.hindawi.com Volume 2018

Diabetes ResearchJournal of

Hindawiwww.hindawi.com Volume 2018

Hindawiwww.hindawi.com Volume 2018

Research and TreatmentAIDS

Hindawiwww.hindawi.com Volume 2018

Gastroenterology Research and Practice

Hindawiwww.hindawi.com Volume 2018

Parkinson’s Disease

Evidence-Based Complementary andAlternative Medicine

Volume 2018Hindawiwww.hindawi.com

Submit your manuscripts atwww.hindawi.com